Release Notes
Also available as:

Behavioral Changes

Behavioral changes denote a marked change in behavior from the previously released version to this version of software. In HDP 2.6.4, behavioral changes affect the following Hadoop components.

Table 1.3. Behavioral Changes

Hortonworks Bug IDApache ComponentApache JIRASummaryDetails
BUG-90722Hive2HIVE-17079LLAP: Use FQDN by default for work submission

Previous Behavior: The default value of LLAP_DAEMON_AM_USE_FQDN is false.

New Behavior: The default value of LLAP_DAEMON_AM_USE_FQDN is now true.

BUG-91117ZeppelinZEPPELIN-1962Queries takes lot of time from Zeppelin Jdbc Interpreter.

Previous Behavior: Hive queries like “show databases” from Zeppelin Jdbc interpreter take a lot of time whereas the same query completes in seconds from Beeline. Multiple thread dump from Hive Jdbc Interpreter process running in Zeppelin Server machine will show a thread performing “SqlCompleter.getDataModelMetadataCompletions” is in RUNNABLE state which confirms the issue.

New Behavior: A patch to disable SQL Completer feature (ZEPPELIN-1962) is available as a workaround. This makes JDBC work with lower latency at the cost of completion functionality. In a future update of Zeppelin, we plan to include a re-write of SQL Completer functionality to improve the performance.

BUG-91290Hive, RangerN/AAdditional ranger hive policies required for INSERT OVERWRITE

Scenario:Additional ranger hive policies required for INSERT OVERWRITE

Previous behavior: Hive INSERT OVERWRITE queries succeed as usual.

New behavior: Hive INSERT OVERWRITE queries are unexpectedly failing after upgrading to HDP-2.6.x with the error:

Error while compiling statement: FAILED: HiveAccessControlException Permission denied: user jdoe does not have WRITE privilege on /tmp/*(state=42000,code=40000)

As of HDP-2.6.0, Hive INSERT OVERWRITE queries require a Ranger URI policy to allow write operations, even if the user has write privilege granted through HDFS policy.

Workaround/Expected Customer Action:

  1. Create a new policy under the Hive repository.

  2. In the dropdown where you see Database, select URI.

  3. Update the path (Example: /tmp/*)

  4. Add the users and group and save.

  5. Retry the insert query.

BUG-91359Hive, Hive2HIVE-17963Hive job creates buckets. Since there are duplicate bucket files, the subsequent validation query (select *) fails to return correct number of records.

Previous Behavior: Setting the property HIVE_EXEC_MOVE_FILES_FROM_SOURCE_DIR could potentially cause runaway tasks to write files to temp directories for non-bloblstore file-systems.

New Behavior: Use HIVE_BLOBSTORE_SUPPORTED_SCHEMES to distinguish blobstores from other file-systems. When moving blobstores or renaming a directory that contains a blobstore, each file is individually moved from the source directory to the destination instead of simply renaming the source directory.

RMP-9153ZeppelinZEPPELIN-1515Support Zeppelin HDFS storage

Previous Behavior: In releases of Zeppelin earlier than HDP-2.6.3, notebooks and configuration files were stored on the local disk of the Zeppelin server.

New Behavior: With HDP-2.6.3+, the default storage is now in HDFS.

Workaround/Expected Customer Action: When upgrading to HDP-2.6.3+ from versions earlier than HDP-2.6.3, perform the steps described in Enabling HDFS Storage for Zeppelin Notebooks and Configuration in HDP-2.6.3+.